Archive for JSM 2010

Vancouver skyline [4]

Posted in pictures, Travel with tags , on August 5, 2010 by xi'an

At Starbucks

Posted in Travel with tags , , , on August 4, 2010 by xi'an

Heard in the queue while waiting for a chai tea latte at Starbucks on the Vancouver waterfront: “…again going to Berlin, Paris, Geneva and London, all for work… Google has a problem in Europe, first it is not a European company, second they are worried about privacy, intellectual property, monopolies, all that…

JSM 2010 [day 3]

Posted in Books, R, Statistics, University life with tags , , , , on August 4, 2010 by xi'an

The same pattern as yesterday occurred, namely that the sessions that most interested me were all together at the 10:30 and 2pm slots, while there was no talk (besides the official ones) after 4pm… After another run around Stanley Park, I took the opportunity of being late for the first session to sit outside and to read the beginning of Francisco Samaniego’s book comparing Bayesian and frequentist inferences. The morning session was SAMSI sum-up about the spatio-temporal program they ran last year. I liked Noel Cressie’s analysis of a huge satellite data problem and the immense dimension reduction it brought. The most promising session of the day was however the afternoon’s Latent space models for network analysis, where Peter Hoff, David Bank and Purnamrita Sarkar gave different perspectives on this quite interesting modelling technique. (I was reflecting during the talks that this could bring a modelling revival of the old French “analyse de données” school, in that it was translating dependencies into distances…) After meeting the other editors of StatProb and grabbing a Revolution Analytic water-bottle at their party, I attended the SBSS mixer where I had a great time (except for the $9 beers) talking to old friends and meeting new ones.

JSM 2010 [day 2]

Posted in Books, pictures, R, Running, Statistics, University life with tags , , , , , , , on August 3, 2010 by xi'an

After a very good early run in Stanley Park, I went to a morning session on new statistical challenges in genetics, but unfortunately could not keep focussed enough (due to a very short night, still being not tuned to Pacific time!) so I ended up chatting with Sid Chib at the Springer booth about the future of R and the drawback of it running too slowly… The second session of the morning I attended was the I.J. Good memorial session (although there were many alternative choices I could have made at the same time!) where Steve Fienberg, Jim Berger, Adrian Raftery and David Banks gave different perspectives on the life and influence of this leading figure. After his work in Bletchley Park along Alan Turing during the war, already using Bayes factors introduced a few years earlier by Harold Jeffreys, I.J. Good contributed very much to the Bayesian revival of the 50’s. (A fact not mentioned this morning is that he was a consultant for 2001: A Space Odyssey!) The afternoon session on Bayesian processing of massive data systems was somehow compulsory since I was talking in this session! While the talks were interestingly diverse, there were however again very people in the room, making me feel the attendance was much lower than last year. As the day ended earlier to let free space to the presidential address, this eventually came as a less exciting day (but left me time for an early evening swim plus two mixers!)…

JSM 2010 [day 1]

Posted in R, Statistics, University life with tags , , , , , , , , , , on August 2, 2010 by xi'an

The first day at JSM is always a bit sluggish, as people slowly drip in and get their bearings. Similar to last year in Washington D.C., the meeting takes place in a huge conference centre and thus there is no feeling of overcrowded [so far]. It may also be that the peripheric and foreign location of the meeting put some regular attendees off (not to mention the expensive living costs!).

Nonetheless, the Sunday afternoon sessions started with a highly interesting How Fast Can We Compute? How Fast Will We Compute? session organised by Mike West and featuring Steve Scott, Mark Suchard and Qanli Wang. The topic was on parallel processing, either via multiple processors or via GPUS, the later relating to the exciting talk Chris Holmes gave at the Valencia meeting. Steve showed us some code in order to explain how feasible the jump to parallel programming—a point demonstrated by Julien Cornebise and Pierre Jacob after they returned from Valencia—was, while stressing the fact that a lot of the processing in MCMC runs was opened to parallelisation. For instance, data augmentation schemes can allocate the missing data in a parallel way in most problems and the same for independent data likelihood computation. Marc Suchard focussed on GPUs and phylogenetic trees, both of high interest to me!, and he stressed the huge gains—of the order of hundreds in the decrease in computing time—made possible by the exploitation of laptop [Macbook] GPUs. (If I got his example correctly, he seemed to be doing an exact computation of the phylogeny likelihood, not an ABC approximation… Which is quite interesting, if potentially killing one of my main areas of research!) Qanli Wang linked both previous with the example of mixtures with a huge number of components. Plenty of food for thought.

I completed the afternoon session with the Student Paper Competition: Bayesian Nonparametric and Semiparametric Methods which was discouragingly empty of participants, with two of the five speakers missing and less than twenty people in the room. (I did not get the point about the competition as to who was ranking those papers. Not the participants apparently!)